upd md splits

README.md

@@ -113,7 +113,11 @@ for item in dataset:
 ```
 
 ### Data Splits
-Train(80%), valid(10%), test(10%)
+|      total      | 36375 |
+| :-------------: | :---: |
+|   train(80%)    | 29100 |
+| validation(10%) | 3637  |
+|    test(10%)    | 3638  |
 
 ## Dataset Creation
 ### Curation Rationale